Self-optimization method and device for monitoring indexes, equipment and medium

Abstract

The invention relates to the technical field of artificial intelligence, and discloses a self-optimization method, device, equipment and medium for a monitoring index, and the method comprises the steps: finding out the difference between a historical index attribute and an existing index attribute based on machine learning through collecting the existing index attribute and the historical index attribute, and carrying out the self-optimization of the historical index attribute; and analyzing the correlation between the historical index attributes and the existing index attributes, generating an index optimization suggestion, establishing an association relationship of a plurality of index attributes, analyzing the change of each index attribute in the association relationship, and generating an index attribute correction and index attribute splitting suggestion. Therefore, the monitoring system realizes intelligent regulation and control, the influence caused by man-made misjudgment and omission is reduced, the accuracy of index attribute regulation and control is improved, index attribute optimization suggestions are continuously given through demand change, an index association relationship and a misinformation feedback mechanism, and the data of the index attributes are continuously matched with the monitoring system, so that when a problem occurs, the accuracy of index attribute regulation and control is improved. The monitoring system finds problems first and makes adaptive repair timely.

Description

Technical Field

[0001] The present invention relates to the fields of artificial intelligence technology and natural language processing technology, and in particular to a self-optimization method, device, equipment and medium for monitoring indicators. Background Art

[0002] Monitoring is an integral part of the software system, which can provide users with monitoring, alarm, notification and other functions. Technical monitoring (CPU usage, memory usage, interface time consumption, etc.) is more general and often does not require developers to participate too much in the formulation. Business monitoring is generally closely related to the current business characteristics and requires developers to customize it.

[0003] In existing products, developers initially create some monitoring indicators based on the business system and preset an alarm threshold. They then continuously observe and adjust the threshold based on analysis. When demand changes or data changes significantly, they manually analyze demand conditions and existing indicators to assess whether indicators and indicator thresholds need to be adjusted. When a production accident occurs or a customer complaint is received, they analyze why the alarm was not issued, whether the evaluation indicators need to be split, and whether the threshold needs to be optimized. The above adjustment directions rely entirely on the developer's experience analysis, and the accuracy of the indicators cannot be guaranteed, resulting in a large number of false alarms. Summary of the invention

[0004] The present invention provides a self-optimization method, device, computer equipment and medium for monitoring indicators to solve the technical problem of completely relying on the experience analysis of developers to adjust thresholds when demand changes or data has significant changes.

[0005] In a first aspect, a self-optimization method for monitoring indicators is provided, comprising:

[0006] Collect existing indicator attributes and historical indicator attributes;

[0007] Based on machine learning, find out the differences between historical indicator attributes and existing indicator attributes, analyze the correlation between historical indicator attributes and existing indicator attributes, and generate indicator optimization suggestions;

[0008] Establish the correlation between multiple indicator attributes, analyze the changes of each indicator attribute in the correlation, and generate indicator attribute modification and indicator attribute splitting suggestions.

[0009] In a second aspect, a self-optimization device for monitoring indicators is provided, comprising:

[0010] The collection module is used to collect the current indicator attributes and historical indicator attributes;

[0011] The optimization module is used to find the differences between historical indicator attributes and existing indicator attributes based on machine learning, analyze the correlation between historical indicator attributes and existing indicator attributes, and generate indicator optimization suggestions;

[0012] The correction module is used to establish the correlation relationship between multiple indicator attributes, analyze the changes of each indicator attribute in the correlation relationship, and generate indicator attribute correction and indicator attribute splitting suggestions.

[0013] In a third aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the self-optimization method of the above-mentioned monitoring indicators when executing the computer program.

[0014] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the self-optimization method of the above-mentioned monitoring indicators are implemented.

[0015] In the scheme implemented by the self-optimization method, device, computer equipment and storage medium of the above-mentioned monitoring indicators, it is possible to collect existing indicator attributes and historical indicator attributes, find out the differences between historical indicator attributes and existing indicator attributes based on machine learning, and analyze the correlation between historical indicator attributes and existing indicator attributes, generate indicator optimization suggestions, establish associations between multiple indicator attributes, analyze the changes in each indicator attribute in the association, and generate indicator attribute correction and indicator attribute splitting suggestions. The monitoring system can achieve intelligent regulation, reduce the impact of human misjudgment and omissions, increase the accuracy of indicator attribute regulation, and continuously give indicator attribute optimization suggestions through demand changes, indicator associations, and false alarm feedback mechanisms, so that the data of indicator attributes continues to match the monitoring system, so that when problems occur, the monitoring system is the first to discover the problem and make adaptive repairs in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0017] Figure 1 It is a schematic diagram of an application environment of a self-optimization method for monitoring indicators in one embodiment of the present invention;

[0018] Figure 2 It is a flow chart of a self-optimization method of monitoring indicators in one embodiment of the present invention;

[0019] Figure 3It is a flowchart of a specific implementation method of a self-optimization method for monitoring indicators in one embodiment of the present invention;

[0020] Figure 4 It is a structural schematic diagram of a self-optimization device for monitoring indicators in one embodiment of the present invention;

[0021] Figure 5 is a schematic diagram of a structure of a computer device in one embodiment of the present invention;

[0022] Figure 6 It is another structural schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] The self-optimization method for monitoring indicators provided by the embodiment of the present invention can be applied in the following aspects: Figure 1 In the application environment, the client communicates with the server through the network. The server can collect existing indicator attributes and historical indicator attributes, find the difference between historical indicator attributes and existing indicator attributes based on machine learning, and analyze the correlation between historical indicator attributes and existing indicator attributes, generate indicator optimization suggestions, establish the association relationship of multiple indicator attributes, analyze the changes of each indicator attribute in the association relationship, and generate indicator attribute correction and indicator attribute splitting suggestions. The monitoring system can realize intelligent regulation, reduce the impact of human misjudgment and omissions, increase the accuracy of indicator attribute regulation, and continuously give indicator attribute optimization suggestions through demand changes, indicator association relationships, and false alarm feedback mechanisms, so that the data of indicator attributes continues to match the monitoring system, so that when problems occur, the monitoring system takes the lead in discovering problems and making adaptive repairs in time. Among them, the client can be but is not limited to various personal computers, laptops, smart phones, tablet computers and portable wearable devices. The server can be implemented with an independent server or a server cluster composed of multiple servers. The present invention is described in detail below through specific embodiments.

[0025] See also Figure 2 As shown, Figure 2 A flowchart of a self-optimization method for monitoring indicators provided by an embodiment of the present invention includes the following steps:

[0026] S10: Collect existing indicator attributes and historical indicator attributes;

[0027] Existing indicator attributes refer to the real-time monitoring indicator attributes collected, and historical indicator attributes refer to the monitoring indicator attributes collected within a preset time period, where the preset time period is set according to specific needs. The indicator attributes include but are not limited to documents, interface documents, indicator-related interface access volume, data changes, code changes, test cases, SQL scripts, Apollo configuration, number of production machine nodes, and the environment involved.

[0028] S20: Based on machine learning, find out the differences between historical indicator attributes and existing indicator attributes, analyze the correlation between historical indicator attributes and existing indicator attributes, and generate indicator optimization suggestions;

[0029] Based on machine learning, historical indicator attributes are compared with existing indicator attributes to find out the differences between them, and the correlation between them is analyzed to obtain the normal numerical rules of indicator attributes and generate indicator optimization suggestions.

[0030] Among them, machine learning algorithms use computational methods to "learn" information directly from data, use limited observed indicator attributes to learn or guess the general laws, and use these laws to predict indicator attributes. In machine learning, it is necessary to preprocess the indicator attributes, select and extract features, select and train models, evaluate and tune models, and finally use the test set to evaluate the performance of the optimized model.

[0031] Specifically, the preprocessing of indicator attributes includes: cleaning, sorting and conversion of indicator attributes. Data cleaning: For indicator attributes, we need to delete unnecessary labels and special characters, as well as duplicate and invalid data;

[0032] Indicator attribute arrangement: merge all indicator attributes into one indicator attribute set, and mark each indicator attribute with a category (necessary indicator attribute or non-essential indicator attribute). 0 can be used to represent non-essential indicator attributes and 1 can be used to represent necessary indicator attributes. Data conversion: convert each indicator attribute into a numerical feature suitable for processing by machine learning algorithms. Technologies such as bag-of-words model or TF-IDF can be used to convert text data into vector representation. Data partitioning: divide the entire indicator attribute set into training set and test set. Usually, we will use most of the indicator attributes for training and a small part for model evaluation. Data balancing: if the number of necessary indicator attributes far exceeds the number of non-essential indicator attributes, consider balancing the data to avoid the model being biased towards predicting necessary indicator attributes. Feature selection: select the most informative features for training. Statistical methods can be used. Data normalization: normalize the features, such as scaling the numerical features to the range of 0-1 so that the model can handle them better.

[0033] Feature selection uses the most relevant features to train the model, or extracts useful features from the raw data through feature extraction algorithms. Feature selection: Univariate feature selection: Calculate the correlation between each feature and the target variable, and select the feature with the highest correlation. Statistical indicators such as chi-square test, Pearson correlation coefficient, etc. can be used; Model-based feature selection: Use machine learning algorithms to train the model and select the most useful features based on the feature importance or coefficient of the model; Filtering feature selection: Filter features based on indicators such as correlation and variance between features; Embedded feature selection: Automatically select features with higher weights or importance during model training. Feature extraction: Text feature extraction: Convert the data with text in the indicator attribute into a numerical feature vector. Common methods include bag-of-words model, TF-IDF (term frequency-inverse document frequency), etc.; Image feature extraction: The data with image in the indicator attribute extracts the numerical features of the image by calculating the color histogram, texture features, edge features, etc. of the image; Speech feature extraction: The data with speech in the indicator attribute extracts the frequency, energy, pitch and other features of the sound from the speech signal. Finally, the purpose of feature selection (filtering, packaging or embedding methods can be used) and feature extraction is to reduce the dimensionality of the feature space, improve the training effect and generalization ability of the model, select and extract the most relevant and informative features, and help the model better understand the patterns and trends in the data, thereby improving the accuracy of classification or prediction.

[0034] Model selection and training: You need to select a machine learning algorithm suitable for the task and use the labeled data to train the model. Task type: First, determine whether the task is classification, regression, or clustering. Classification tasks are used to classify data, regression tasks are used to predict continuous values, and clustering tasks are used to divide data into different clusters. Different task types are suitable for different algorithms; Data scale: Consider the scale of your data set. For small-scale data sets, you can choose some statistical-based algorithms (such as Naive Bayes), while for large-scale data sets, you can choose some more efficient algorithms (such as support vector machines or deep learning algorithms); Feature attributes: Analyze feature attributes. If the feature is a continuous value, you can choose a regression algorithm; if it is a discrete value, you can choose a classification algorithm. You can also choose a suitable algorithm based on the number of feature attributes. For example, decision trees are suitable for processing data with multiple attributes; Data quality: Consider the quality of your data. If there is noise or missing values ​​in the data, some more robust algorithms (such as random forest algorithm models) may be more suitable for handling the above situations.

[0035] Once the algorithm suitable for the task is determined, the model can be trained using the labeled data. This can be done by following the steps below: Split the dataset into training and test sets: Divide the dataset into two parts, one for training the model and the other for evaluating the performance of the model.

[0036] Feature engineering: preprocessing and feature extraction of data, which may include data cleaning, standardization, feature selection, feature conversion and other operations;

[0037] Model training: Input the preprocessed data into the selected machine learning algorithm and use the training set to train the model. This usually involves adjusting the algorithm's hyperparameters to achieve better performance; Model evaluation: Use the test set to evaluate the performance of the model. Common evaluation indicators include accuracy, precision, recall, F1 score, etc.

[0038] Model evaluation and tuning: It is necessary to evaluate the performance of the trained model on new data, and tune the model based on the evaluation results to improve its performance. The purpose of evaluating the indicator properties of the trained model on new data is to determine whether the model has generalization ability, that is, the prediction accuracy on unseen data. Divide the data set: Divide the data set into training set, validation set and test set. The training set is used for model training, the validation set is used to tune the model's hyperparameters, and the test set is used to finally evaluate the model performance; Select evaluation indicators: Select appropriate evaluation indicators according to the task type. For example, commonly used indicators in classification tasks include accuracy, precision, recall rate, F1 score, etc.; commonly used indicators in regression tasks include mean square error (MSE), mean absolute error (MAE), coefficient of determination (R2), etc.; Model evaluation: Use the validation set to evaluate the performance of the model on new data. According to the selected evaluation indicator, calculate the prediction results of the model on the validation set and compare them with the true label to obtain the value of the evaluation indicator; tune the hyperparameters: adjust the hyperparameters of the model according to the evaluation results on the validation set. You can use different hyperparameter combinations, compare their performance on the validation set, and select the best performing combination; model retraining: after determining the best hyperparameter combination, use the data from the training set and validation set to retrain the model, so that more data can be used to train the model and improve the performance of the model; model evaluation: use the test set to evaluate the performance of the final optimized model to judge the performance of the model on unknown data.

[0039] Finally, in the real-time prediction process, the data of indicator attributes is obtained in real time, such as the number of visits to the indicator-related interface, the amount of data changes, the indicator time range, and the newly added channels / statuses. Then, this data is input into the trained model and predicted through the model. The model will judge the data fluctuation trend of the indicator attributes, such as rising, falling, or flat, based on the historical indicator attributes and other features.

[0040] Based on the prediction results of the model, real-time decisions can be made. For example, if the model predicts that the data of an indicator attribute will increase, we can decide to add data channels or analyze whether the indicator attribute needs to be split. If the model predicts that the data of an indicator attribute will decrease, we can decide to close the data channel and recommend the indicator to go offline.

[0041] Through real-time prediction and real-time decision-making, we can more flexibly respond to data changes in indicator attributes. However, it should be noted that the prediction results of the model may have a certain degree of error, so we still need to be cautious about the prediction results and make decisions based on other information.

[0042] S30: Establishing association relationships among multiple indicator attributes, analyzing changes in each indicator attribute in the association relationship, and generating indicator attribute modification and indicator attribute splitting suggestions.

[0043] Based on the above training model, the association relationship of multiple indicator attributes is found, and multiple indicator attributes are associated based on the association relationship of multiple indicator attributes. The changes of each indicator attribute in the association relationship are analyzed. Based on the association relationship of multiple indicator attributes, the changes of each indicator attribute are analyzed after the changes of each indicator attribute. The accuracy of the real-time indicator attribute is determined by comparing the prediction result with the real-time indicator attribute. If the accuracy of the real-time indicator attribute is low, a correction plan for the indicator attribute is given based on the prediction result. If the accuracy of the real-time indicator attribute is high, a suggestion for splitting the indicator attribute is given based on the prediction result.

[0044] In this embodiment, when the amount of data or the amount of interface access increases greatly, the number of user accesses increases, and the index splitting optimization is performed according to different users. The amount of data change in the current index table and the amount of interface access related to the index, the increase in the amount of data or the amount of interface access means that the number of users accessing has increased. Different levels of data often have different monitoring index focus points. The index is split and optimized according to different user types, and correction and splitting suggestions are given by analyzing the changes in related indicators.

[0045] like Figure 3 As shown, after the steps of establishing the association relationship of multiple indicator attributes, analyzing the changes of each indicator attribute in the association relationship, and generating indicator attribute correction and indicator attribute splitting suggestions, the method further includes: S40, based on analyzing the correlation between historical indicator attributes and existing indicator attributes, monitoring whether the existing indicator attributes are false alarms; if yes, S41, analyzing the causes and frequency of false alarms, comprehensively analyzing and giving optimization and splitting and threshold setting suggestions, and establishing a false alarm feedback mechanism; if no, executing the step of monitoring whether the existing indicator attributes are false alarms. When the target indicator attribute continuously misreports the target number of times, the target indicator attribute is taken offline. Analyze the overall optimization suggestions based on the optimization suggestions, indicator attribute correction and indicator attribute splitting suggestions, and the false alarm feedback mechanism.

[0046] Before the step of analyzing the causes and frequencies of false alarms, comprehensively analyzing and giving suggestions for optimization and splitting as well as threshold setting, and establishing a false alarm feedback mechanism, the method further includes: monitoring whether the number of false alarms is greater than one: if yes, executing the step of analyzing the causes and frequencies of false alarms, comprehensively analyzing and giving suggestions for optimization and splitting as well as threshold setting, and establishing a false alarm feedback mechanism; if no, executing the step of monitoring whether the existing indicator attributes are false alarms.

[0047] Specifically, monitor whether the indicator attributes are false alarms and analyze the reasons for false alarms. There may be differences. It is difficult to analyze optimization points for a single false alarm. Through the false alarm feedback mechanism, analyze the reasons and frequency of multiple false alarms, and give optimization and splitting and threshold setting suggestions through comprehensive analysis. Of course, for continuous false alarms, it is possible to give suggestions for indicator offline.

[0048] It can be seen that in the above scheme, intelligent regulation is performed on the monitored indicator attributes. By collecting existing indicator attributes and historical indicator attributes, based on machine learning, the differences between historical indicator attributes and existing indicator attributes are found, and the correlation between historical indicator attributes and existing indicator attributes is analyzed, and indicator optimization suggestions are generated. The association relationship of multiple indicator attributes is established, and the changes of each indicator attribute in the association relationship are analyzed, and indicator attribute correction and indicator attribute splitting suggestions are generated. The monitoring system can achieve intelligent regulation, reduce the impact of human misjudgment and omissions, and increase the accuracy of indicator attribute regulation. Through demand changes, indicator association relationships, and false alarm feedback mechanisms, indicator attribute optimization suggestions are continuously given, so that the data of indicator attributes continues to match the monitoring system, so that when problems occur, the monitoring system is the first to discover the problem and make adaptive repairs in time.

[0049] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0050] In one embodiment, a self-optimization device for monitoring indicators is provided, and the self-optimization device for monitoring indicators corresponds one-to-one to the self-optimization method for monitoring indicators in the above embodiment. Figure 4 As shown, the self-optimization device for monitoring indicators includes a collection module 101, an optimization module 102, and a correction module 103. The detailed description of each functional module is as follows:

[0051] The collection module 101 is used to collect the current indicator attributes and the historical indicator attributes;

[0052] The optimization module 102 is used to find out the differences between the historical indicator attributes and the existing indicator attributes based on machine learning, and to analyze the correlation between the historical indicator attributes and the existing indicator attributes, and to generate indicator optimization suggestions;

[0053] The correction module 103 is used to establish an association relationship among multiple indicator attributes, analyze changes in each indicator attribute in the association relationship, and generate indicator attribute correction and indicator attribute splitting suggestions.

[0054] In one embodiment, the device is further configured to:

[0055] Based on the analysis of the correlation between historical indicator attributes and existing indicator attributes, monitor whether the existing indicator attributes are false alarms; if yes, analyze the causes and frequency of false alarms, make comprehensive analysis to provide optimization, splitting and threshold setting suggestions, and establish a false alarm feedback mechanism; if no, execute the steps of monitoring whether the existing indicator attributes are false alarms.

[0056] In one embodiment, the device is further configured to:

[0057] When the target indicator attribute continues to report the target number of times incorrectly, the target indicator attribute is taken offline.

[0058] In one embodiment, the device is further configured to:

[0059] Monitor whether the number of false alarms is greater than one: if yes, perform steps of analyzing the causes and frequencies of false alarms, comprehensively analyzing to provide optimization and splitting and threshold setting suggestions, and establishing a false alarm feedback mechanism; if no, perform steps of monitoring whether existing indicator attributes are false alarms.

[0060] In one embodiment, the device is further configured to:

[0061] Analyze the overall optimization suggestions based on optimization suggestions, indicator attribute correction and indicator attribute splitting suggestions, and false alarm feedback mechanism.

[0062] The present invention provides a self-optimization device for monitoring indicators. By collecting existing indicator attributes and historical indicator attributes, based on machine learning, the differences between historical indicator attributes and existing indicator attributes are found, and the correlation between historical indicator attributes and existing indicator attributes is analyzed, indicator optimization suggestions are generated, and association relationships between multiple indicator attributes are established. Changes in each indicator attribute in the association relationship are analyzed, and indicator attribute correction and indicator attribute splitting suggestions are generated. The monitoring system is enabled to achieve intelligent regulation, reduce the impact of human misjudgment and omissions, increase the accuracy of indicator attribute regulation, and continuously give indicator attribute optimization suggestions through demand changes, indicator association relationships, and false alarm feedback mechanisms, so that the data of indicator attributes continues to match the monitoring system, so that when problems occur, the monitoring system is the first to discover the problem and make adaptive repairs in a timely manner.

[0063] For the specific definition of the self-optimization device of the monitoring index, please refer to the definition of the self-optimization method of the monitoring index in the above text, which will not be repeated here. Each module in the self-optimization device of the above monitoring index can be implemented in whole or in part by software, hardware and a combination thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0064] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external client via a network connection. When the computer program is executed by the processor, it realizes the functions or steps of a self-optimization method for monitoring indicators on the server side.

[0065] In one embodiment, a computer device is provided. The computer device may be a client, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server via a network connection. When the computer program is executed by the processor, it realizes the functions or steps of the client side of a self-optimization method for monitoring indicators

[0066] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0067] Collect existing indicator attributes and historical indicator attributes;

[0068] Based on machine learning, find out the differences between historical indicator attributes and existing indicator attributes, analyze the correlation between historical indicator attributes and existing indicator attributes, and generate indicator optimization suggestions;

[0069] Establish the correlation between multiple indicator attributes, analyze the changes of each indicator attribute in the correlation, and generate indicator attribute modification and indicator attribute splitting suggestions.

[0070] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0071] Collect existing indicator attributes and historical indicator attributes;

[0072] Based on machine learning, find out the differences between historical indicator attributes and existing indicator attributes, analyze the correlation between historical indicator attributes and existing indicator attributes, and generate indicator optimization suggestions;

[0073] Establish the correlation between multiple indicator attributes, analyze the changes of each indicator attribute in the correlation, and generate indicator attribute modification and indicator attribute splitting suggestions.

[0074] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can refer to the relevant descriptions on the server side and the client side in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.

[0075] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0076] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0077] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A self-optimization method for monitoring indicators, characterized in that: include: Collect existing indicator attributes and historical indicator attributes; Based on machine learning, find out the differences between historical indicator attributes and existing indicator attributes, analyze the correlation between historical indicator attributes and existing indicator attributes, and generate indicator optimization suggestions; Establish the correlation between multiple indicator attributes, analyze the changes of each indicator attribute in the correlation, and generate indicator attribute modification and indicator attribute splitting suggestions.

2. The self-optimization method for monitoring indicators according to claim 1, characterized in that: After the steps of establishing association relationships of multiple indicator attributes, analyzing changes of each indicator attribute in the association relationships, and generating indicator attribute modification and indicator attribute splitting suggestions, the method further includes: Based on the analysis of the correlation between historical indicator attributes and existing indicator attributes, monitor whether the existing indicator attributes are false alarms; Yes, analyze the causes and frequency of false alarms, make comprehensive analysis to provide optimization, splitting and threshold setting suggestions, and establish a false alarm feedback mechanism; No, perform the steps to monitor existing indicator attributes for false positives.

3. The self-optimization method for monitoring indicators according to claim 2, characterized in that: After the steps of analyzing the causes and frequencies of false alarms, comprehensively analyzing and providing optimization and splitting and threshold setting suggestions, and establishing a false alarm feedback mechanism, the method further includes: When the target indicator attribute continues to report the target number of times incorrectly, the target indicator attribute is taken offline.

4. The self-optimization method for monitoring indicators according to claim 2, characterized in that: Before the step of analyzing the causes and frequencies of false alarms, comprehensively analyzing and providing optimization and splitting and threshold setting suggestions, and establishing a false alarm feedback mechanism, the method further includes: Monitor whether the number of false alarms is greater than one: Yes, perform the steps of analyzing the causes and frequency of false alarms, making comprehensive analysis to provide optimization and splitting and threshold setting suggestions, and establishing a false alarm feedback mechanism; No, perform the steps to monitor existing indicator attributes for false positives.

5. The self-optimization method for monitoring indicators according to claim 2, characterized in that: After the steps of analyzing the causes and frequencies of false alarms, comprehensively analyzing and providing optimization and splitting and threshold setting suggestions, and establishing a false alarm feedback mechanism, the method further includes: Analyze the overall optimization suggestions based on optimization suggestions, indicator attribute correction and indicator attribute splitting suggestions, and false alarm feedback mechanism.

6. The self-optimization method for monitoring indicators according to claim 1, characterized in that: The indicator attributes include but are not limited to documents involved, interface documents, code changes, test cases, SQL scripts, Apollo configuration, number of production machine nodes, and the environment involved.

7. The self-optimization method for monitoring indicators according to claim 6, characterized in that: When the amount of data or the number of interface visits increases significantly, the number of user visits increases, and indicators are split and optimized according to different users.

8. A self-optimization device for monitoring indicators, characterized in that: include: The collection module is used to collect the current indicator attributes and historical indicator attributes; The optimization module is used to find the differences between historical indicator attributes and existing indicator attributes based on machine learning, analyze the correlation between historical indicator attributes and existing indicator attributes, and generate indicator optimization suggestions; The correction module is used to establish the correlation relationship between multiple indicator attributes, analyze the changes of each indicator attribute in the correlation relationship, and generate indicator attribute correction and indicator attribute splitting suggestions.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the self-optimization method for monitoring indicators as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the self-optimization method for monitoring indicators as described in any one of claims 1 to 7 are implemented.

Images